Generally speaking, linearity is an essential feature of an accurate measuring apparatus. More specifically, when a signal detected by a measuring apparatus is amplified by a certain gain, the amplified output signal must be proportional to the detected signal. If this proportionality is not respected and no correction is made, the resulting signal is likely to be erroneous and lead to wrongful interpretation.
In the area of non-destructive testing (NDT) of materials and structures, a so-called “phase array” ultrasonic technique uses n channels including, for example, an array of n ultrasonic transducers. The n ultrasonic transducers are excited according to a given focal law delay profile to ultrasonically scan the tested material or structure. Echoes from defects or flaws present in the tested material or structure are detected by the n ultrasonic transducers to produce corresponding separate channel signals. An average of the channel signals is calculated prior to processing thereof.
Those of ordinary skill in the art will appreciate that saturation of at least one of the n channels will cause non-linearity of the measuring apparatus. More specifically, the calculated average will be erroneous and not representative of the reality.